¿Los toques de queda para menores de edad disminuyen la tasa de homicidio? Evidencia de Cali (Colombia)

Using the method of difference-in-difference (DD), we estimate the average impact on the homicide rate of a Curfew for Juveniles (CFJ) that was implemented (in 2012) in some communes of Cali (Colombia), one of the most violent cities in the world. We find that the policy did not cause a negative effect on the homicide rate contrary to the objective of the policy. This result is robust to different sample specifications defined to capture local effects —continuous (spillovers) or discontinuous—, and biases in the policy design. We study DD with heterogeneous effects in which neighborhoods that had the presence of higher socioeconomic income did reduce the homicide rate even though the average net effect for these types of neighborhoods was not different from zero. We also find that the CFJ neutralized the effects of other public complementary interventions. We rationalize the results using the economics of crime framework, where we discuss the possible incentives a CFJ could generate for delinquents as well as non-delinquents that explains some failures of the policy in its design.